spheres, so that women and men benefit equally, and inequality is not perpetuated. ... women against poverty and we need to examine the quality of jobs created. .... policies led to the raising of the retirement age for women in some countries ...
EUROMOD WORKING PAPER SERIES
EUROMOD Working Paper No. EM8/04 EU ACTION ON SOCIAL INCLUSION AND GENDER MAINSTREAMING A. B. Atkinson and D. Meulders December 2004
Editorial note This paper was first drafted in 2002. Although now out-of-date in the sense that since it was written a new wave of National Action Plans has been produced, it is felt that the analysis is of interest. Not only has it not been duplicated elsewhere, but the issues raised remain relevant.
EU Action on Social Inclusion and Gender Mainstreaming1 A B Atkinson and D Meulders
Introduction 1 The NAP Objectives and Contents from a Gender Mainstreaming Perspective 2 Analysis of the Gender Dimension in the National Plans on Social Inclusion 3 Gender Mainstreaming, Policy and the Role of EUROMOD Conclusions
Introduction As embodied in the Treaty of Amsterdam, equality of treatment between women and men is a prominent part of the EU agenda. Member States have agreed that a gender equality perspective should be integrated into all policy areas, an approach referred to as “gender mainstreaming”. In July 1997, the United Nations Economic and Social Council (ECOSOC) defined the concept of gender mainstreaming as follows: "Mainstreaming a gender perspective is the process of assessing the implications for women and men of any planned action, including legislation, policies or programmes, in any area and at all levels. It is a strategy for making the concerns and experiences of women as well as of men an integral part of the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres, so that women and men benefit equally, and inequality is not perpetuated. The ultimate goal of mainstreaming is to achieve gender equality." Different steps can be identified in a gender mainstreaming process: 1. Analysis of the relative situation of women in the field of action addressed by the measure or policy proposed. This implies data collection by sex and an analysis of the explanatory factors of the observed differences. 2. Estimation of the differentiated impact of the proposal by gender 3. If a differentiated impact is found, then analysis is required of its causes, justification and of the necessary corrections to improve the policy. This paper is concerned with the application of gender mainstreaming to the National Action Plans on Social Inclusion (NAPincl), which are the outcome of the new stage in the development of the European social agenda agreed at the Lisbon and Nice European Councils in 2000. In June 2001, EU Member States submitted their first National Action Plans. These Plans were evaluated in the Joint Report on Social Inclusion (European Commission, 2002) and consideration has now been given to the form of the next round of NAPincl due in June 1
This paper was written as part of the MICRESA (Micro Analysis of the European Social Agenda) project, financed by the Improving Human Potential programme of the European Commission (SERD-2001-00099).
2003. In November 2002, Commissioner Anna Diamantopoulou, addressing the European Parliament, described the aim of consolidating the objectives agreed at Nice with “two small but important changes”. The one that concerns us in the present paper is “the strengthening of the gender dimension of the objectives”. As it was put by the Social Protection Committee in its submission to the Council Meeting of 3 December 2002, Member States are being asked to “underline the importance of mainstreaming equality between men and women in all actions ... by taking into account the gender perspective in the identification of challenges, the design, implementation and assessment of policies and measures, the selection of indicators and targets and the involvement of stakeholders”. The aim of this paper is to examine the treatment of gender mainstreaming in the first round of National Action Plans on Social Inclusion and to assess how far EUROMOD, the European-wide tax-benefit model, can be used to advance the objectives described above. As such it is intended to be an input into developing a Europe-wide capacity for policy analysis.
1 The NAP Objectives and Contents from a Gender Mainstreaming Perspective The National Action Plans on Social Inclusion (referred to as NAPincl) are designed in the context of agreed European objectives. Four Objectives have been agreed, the first of which is subdivided into two: Objective 1 To facilitate participation in employment and access by all to resources, rights, goods and services, 1.1 Facilitating participation in employment, 1.2 Facilitating access to resources, rights, goods and services for all, Objective 2 Prevention of the risks of exclusion, Objective 3 Actions to help the most vulnerable, Objective 4 Mobilisation of all relevant actors. In this section, we examine each of these in turn to see what the detailed objectives reveal about the treatment of gender. 1. To facilitate participation in employment and access by all to resources, rights, goods and services 1.1. Facilitating participation in employment In its diagnosis of the determinants of poverty, the Joint Report gives particular emphasis to the role of macro-economic factors, and specifically to the overall employment rate. A core thesis is that “the major cause of exclusion is the lack of employment” (page 109). Employment is also seen as a key mechanism for gender equality. The Danish NAP notes “Danish social policy supports a very high activity rate for men and women alike” (page 8). The German NAPincl states that it is “trying to encourage women to enter the labour market. This accords with the genuine needs of a growing number of women and can at the same time help to reduce the poverty risk of families and lone parents” (page 7). The Swedish NAPincl states on its first page “one of the main tasks of employment policy is to avoid a gender-segregated labour market and contribute in other ways to greater equality between men and women” (page 1).
2
Common features of European labour markets are indeed the lower employment rates of women, their higher rate of part-time work, and their lower quality of jobs. All three are important. The Lisbon and Stockholm targets talk about employment rates but these have to be expressed in full time equivalents in order that employment can realistically protect women against poverty and we need to examine the quality of jobs created. The reduction in poverty risk to which the German NAP refers can only be achieved if the jobs ensure an adequate level of earnings. In their NAPincl, a number of member States cross-refer to the National Action Plans for Employment (the Swedish Employment Plan as attached to their NAPincl). Reference is made to a range of measures, including statutory requirements regarding public employment. (Although little attention is paid to the possible negative impact on women’s employment of the down-sizing of public employment, contracting-out of services, and privatisation of public enterprises.) Individual Member States refer to specific measures. In the field of active labour market policies, the Irish NAPincl refers to the implementation of recommendations by the Report on Access by Women to the Labour Market (page 19). Both Ireland the UK emphasise the role of childcare programmes. One of the routes identified in the NAP towards the promotion of access to stable and quality employment is the provision of training and the putting in place, for those in the most vulnerable groups in society, pathways towards employment. Experience has however shown that the training programs implemented outside companies have often repeated the segregation observed within the labour market by tending to direct women towards activities they have been accustomed to carrying out at home, such as child care, sewing, cooking, cleaning and so on, all of which are types of training which rarely lead to employment. As has been pointed out by Rees (1995), “Education and training funded by the European Commission through the European Social Fund (Lefebvre, 1992) and the Task Force Human Resources, Education, Training and Youth programs (COMETT, FORCE, LINGUA, PETRA, ERASMUS, TEMPUS) (Rees, 1993) have been found to reinforce gendered patterns of participation and segregation, arguably contributing to the skills gap between men and women” (Rees, 1995). In the case of in-company training schemes, various studies have shown that women are less likely to benefit for four reasons: - The selection criteria applied by the employers, who invest less in their female staff; - The times, schedules and locations chosen for the organization of such training schemes, which are often difficult to square with women's other responsibilities; - Atypical workers (such as those doing part-time work or temporary jobs), who are primarily women, do not generally have access to training programs; - These programs are rarely implemented in small and medium enterprises where women are concerned. As quoted in the Belgian NAP (page 22) “training programmes mainly benefit young people, people at work, men, white collars, civil servants”. The problem is recognised in a number of the NAP. The Austrian government attaches considerable importance in its NAPincl to the promotion of education among women, particularly in technology: “by improving the skills of women Austria is tackling one of the root causes of why women are disadvantaged on the
3
labour market” (page 11). The Portuguese NAP refers to the need to eliminate gender inequalities in access to the Information Society (page 28). We suggest that this aspect should be addressed by every country in their NAP, indicating how they intend to insure an equal participation of women to continuous training and, more widely, how they plan to encourage the participation of atypical workers (part time, temporary etc.) in training programmes). Data should be given concerning the relative rate of participation by men and women for total employment, temporary and part time workers, by training orientation. The second strand is the development of policies to promote the reconciliation of work and family life, including the issue of child- and dependent care. This aspect is of high importance for women’s labour force participation. Different countries announce in their NAP policies in order to improve child care provision: tax deduction of costs, premiums, and increase in the availability of places. Among the aspects on which indicators are needed are the availability, the price and the quality of child care facilities, the number of places for children less than 3 years old in the public and the private sector, and their price, the number of children in the pre- compulsory school system, the opening hours of schools, the availability of care for people searching a job, and the extent of tax deductibility of child care. The third strand is the use of the opportunities for integration and employment provided by the social economy. Women are particularly concerned by social economy. The development of neighbourhood services can improve considerably their quality of life and the opportunities for work/life balance. Women are the bulk of workers in social economy, due to the traditionally 'feminine' character of the services referred to in connection with everyday life. This might be seen as a positive aspect, since one might imagine that the majority of the jobs created to satisfy these needs would be for women; obviously it all depends upon the status, remuneration and social protection they would be granted. Most of the work devoted to social economy have underlined the necessity to insure the quality of employment in this sector where social and legal protection are difficult to implement in a relation between two individuals (not a firm and a worker), the danger of developing new forms of “slavery” has been underpinned by different author stressing risks of low pay and inadequate social protection. It is clear that if the jobs created do not enjoy the “classical” status of salaried workers, and if they are limited as to time or reserved for certain categories of people, they will lead to a new kind of atypical employment which will be in direct competition with certain sectors of the commercial economy by penalizing both women working in those sectors and those taking on the new jobs. The quality of employment is in this field a major concern including an important gender dimension that does not appear in the NAP. 1.2. Facilitating access to resources, rights, goods and services for all Employment is one key policy instrument; the second main thrust of policy is via social protection. European social security systems can exert a discriminatory effect on women through two routes. First, these systems do not ensure adequate protection for workers in atypical circumstances, among whom women are over-represented: such persons are often deprived of proper protection, mainly with respect to unemployment and pensions. The proportion of women entitled to some form of assistance is systematically lower than that of men (26% as
4
opposed to 34% in the EU as a whole in 1995). The conditions governing the granting of such benefits very often prove to be an indirect source of discrimination between men and women out of work. In the case of unemployment benefits (insurance principle), women often find it difficult to comply with the minimum contribution requirements, since their occupational profiles are more often marked by interruptions than their male counterparts, and many of them work pan-time. Career breaks, part-time work and temporary work are the source of this inequality with respect to unemployment benefits. Second, the absence of individual entitlements and the proliferation of derived rights forces women into relationships of dependence tinged with insecurity, which discourage them from entering the official labour market and results in inequalities between the benefits received according to family circumstances. Europe’s taxation and social security systems contain various “inactivity traps” which were deliberately set and still exist: derived rights in social protection systems are an incentive for women to remain at home or to engage in undeclared work, since marriage automatically confers on a spouse the healthcare and pension benefits enjoyed by the partner. The advantages granted to the home-based spouse by the tax system, namely due to dependents’ allowances, as well as all the allowances linked to inactivity, such as career breaks, constitute inactivity traps that affect a far larger number of persons of working age than the unintentional traps that are currently being singled out for corrective action (Jepsen, Meulders and Plasman, 1997). As to the social aid associated with unemployment (relief principle), married women and those living with a man are at a disadvantage with regard to the means conditions imposed, since they are dependent on their husband's income. However, the disadvantageous situation of women is emphasized by another most important element: the modulation of the replacement rate according to the family situation. An unemployed person can in some cases claim additional benefit for a dependent spouse, and in view of the labour market situation of women (part-time, temporary employment), this non-individualization of unemployment insurance has the potential to act as a disincentive for women to engage in formal labour market participation. The gender dimension of social protection is not systematically addressed in the 2001 NAP. The Joint Report notes that the Nordic countries emphasise the contribution to gender equality of social policy systems based on individualised rights. It would be interesting to examine the contribution of such system-differences to variation in gender inequality across Member States. Has the move towards family-based means-tested social assistance in the UK, away from individual social insurance, increased gender differences? Has the move to private pensions, with defined contribution formulae, disadvantaged women? Promotion of equal policies led to the raising of the retirement age for women in some countries (Belgium, Italy), however less attention has been paid to the ability of women to get an independent pension after atypical work history. Many Member States refer to social policies that differentially impact on women and men without expressly identifying this. The German long-term care insurance provides a good example. The provision of benefits in cash or kind serves to relieve the burden on carers, who are predominantly women. Outpatient care, for instance, makes “it possible in many cases for care-dependant persons to be looked after at home despite their carer’s employment status” (page 17 of the German NAP). Put in reverse, the outpatient care allows the carers to take up
5
employment while continuing to look after their relatives. In the same way, the UK is providing better support for carers. According to the UK NAPincl (page 21) of 2 million carers, at least 1.5 million are women. According to the Irish NAPincl, 79% of the recipients of the Carer’s Allowance were women (page 17). A second example is child benefit. Not only does child benefit play a key role in preventing poverty among families with children, but also it can be expected to differentially benefit women, particularly lone parents. This can indeed have a continuing impact, as illustrated by the example of life pensions in Greece for mothers with 4+ grown-up children (page 28). Finally, in the Belgian NAP evidence is presented of gender bias in access to culture, sport in leisure. If this is the case in other countries, mainstreaming should lead to a gender questioning of access to these items
2 Analysis of the Gender Dimension in the National Plans on Social Inclusion Member States had no more than 6 months to prepare their NAPincl, and there was limited time for co-ordination. Member States interpreted their brief in different ways. A number of countries provided general accounts of their approach to social protection; others discussed in detail recent legislative changes. Some counties attempted an extensive analysis of social exclusion; others listed a set of general concerns. The NAPincl had however in common the lack of a comprehensive analysis of the gender dimension of social exclusion. The absence of such an integrated approach was disappointing. The National Action Plans on Employment provided a precedent and in this respect it is surprising that the work done on women’s employment issues in the employment field did not impact more on the analysis of social exclusion. Gender mainstreaming has been adopted by the European Union because women are economically disadvantaged, particularly with regard to the labour market. This is a theme that recurs in the NAP, but which is not systematically investigated. Some Member States give more attention than others. The Southern and Nordic countries stand out. This is an area where Member States can learn from the good practice of others. The following sentences from the Joint Report summarise the concerns about the lack of systematic analysis: • Belgium: “The gender dimension in the NAPincl is taken on board but with limited visibility” (page 89). • Greece: “While the intention to adopt a gender mainstreaming approach is clearly expressed, the NAPincl lacks a comprehensive strategy based on a systematic analysis of the problems” (page 108). • Ireland: “There is no gender analysis in the main challenges” (page 126). • Netherlands: “There is little attention for gender mainstreaming in this NAPincl” (page 145).
6
•
• •
Austria: “A consistent gender mainstreaming approach and indications, how the specific problems ... will be addressed, are covered insufficiently and too generally in the NAP” (page 151). Sweden: “There is gender mainstreaming in the structure itself. This might be why there are rather few specific references to gender issues” (page 169). United Kingdom: “its treatment of gender mainstreaming is patchy” (page 176).
It should be noted that gender disadvantage in most cases affects women. Even where they are less heavily represented, as with the homeless, the consequences are more severe. But there are some situations in which men are disadvantaged. In particular, the Portuguese NAP emphasises the need for working men to be able to participate fully in family life. Part 4 of the Joint Report deals with “Promoting Equality between Women and Men” and concludes that, despite the stress at the Nice European Council on the importance of gender mainstreaming, this is not implemented consistently across the NAPincl: “Unfortunately, whilst all Member States mention some gender issues, very few mainstream equality between men and women consistently across their Plans – from the identification of the challenges, through the overall strategy, to the designing and monitoring of detailed measures” (2002, page 74). These different aspects are discussed below. Comparable Indicators of Social Inclusion Given that the NAPincl of the Member States did not provide such a systematic analysis, it is important to develop at the EU-level indicators that are comparable across countries and where the gender differences can be examined. The research of Eurostat using the European Community Household Panel (ECHP)2 has already served to bring out the gender differences along a number of dimensions. Figure 1 shows the relative incidence of low pay in 1996 for women and men as found in the Low pay is defined as having a monthly wage less than 60% of the country’s median wage. In all countries apart from Denmark and Greece women are over-represented among the low earners by more than 50%. The scope is limited to those workers working at least 15 hours a week, but the figures reflect the fact that hours of work vary. (The source is Marlier and Ponthieux, 2000, page 6; figures for 1997 are given for some countries in the Joint Report, page 185.) This is one reason that women are disproportionately represented, but it is not the only reason, as is well known from studies of wage differences. The same applies to the poverty rates shown in Figure 2, where the poverty line is taken as 60% of the median equivalised income per person in each Member State. (The sources are Mejer and Siermann, 2000, page 4 and Mejer and Linden, 2000, page 3.) Women have a higher poverty rate than men in all countries except Spain. This is not only true during working age but also persists into old age (except again in Spain). Women are more likely to live in persistent poverty: i.e. to be in financial poverty for each of the three years 1994, 1005 and 1996. In Ireland the persistent poverty rate for women is double that of men (off the chart).
2
The ECHP is a panel survey based on a standardized questionnaire that involves interviewing a representative sample of households in each country, covering a wide range of topics, including income, health, education, housing and employment. The first wave was conducted in 1994 in the then 12 EU Member States, since then Austria (1995) and Finland (1996) have joined the project. Sweden is not participating. See Eurostat (1999).
7
In measuring poverty, the hypothesis is that resources are shared equally among the members of a household: poor individuals belong to poor households. Certain studies have underlined the importance of power in the distribution of earnings and consumption within households. Recent empirical research has confirmed that the allocation of resources among family members depends on age and gender. Studies conducted in Asia and Africa have emphasised the high rates of female mortality due to the systematic deprivation of resources suffered by women within households (not necessarily poor ones - see Sen, 1989). In one of the few studies on distribution within American households, Lazear and Michael (1988) reveal that the division of consumption between parents and children alters radically with the number of adults in the household: as the number of adults rises, the children’s share declines proportionately. Furthermore, access to paid work considerably changes the share-out of power within the household. On average, women earning an income have more decisionmaking power within the family, and this share of the power increases in step with their earnings (England and Kilbourne, 1990). Findlay and Wright (1996) have assessed, for Italy and the United States, the way in which an unequal distribution of resources within households affected the measurement of poverty. They conclude that the conventional methods of gauging poverty within households lead to a systematic underestimation of female poverty and an overestimation of male poverty. One important issue is the change over time in gender disadvantage. This is discussed in some of the NAPincl. The Greek NAP is particularly interesting. It first draws attention to the “increase in the last few years in the number of women participating in the labour market and higher education. Three-quarters of net inflows into employment in 1990-1999 were women” (page 25). But it also notes that the problems of women may be increasing as a result of divorce and family break-up. In terms of earnings, according to the Irish NAPincl, there has been a narrowing of the gap in hourly earnings, which in 1997 was 15.5% across all economic sectors (page 19). This may have been further reduced by the introduction of the National Minimum Wage. In the UK, the introduction of a National Minimum Wage, benefiting some 1.3 million workers from April 1999; is considered to have had this effect: “in narrowing the gender pay gap the minimum wage has had the greatest effect on women’s pay since the Equal Pay Act 1970” (page 8). The dynamics of gender disadvantage seem a topic of high priority in future NAPincl. Multiple Dimensions of Social Exclusion One of the main thrusts of the social inclusion agenda has been to widen the definition of disadvantage beyond that of financial poverty and income inequality. This brings us to the interesting but little discussed question as to whether the same gender disadvantage is found with other dimensions. What are the implications of the wider social exclusion agenda for the gender issue? The wider approach to social exclusion has been made concrete in the set of social indicators agreed at the Laeken Council in December 2001 (see Social Protection Committee, 2001). Table 1 sets out the ten primary indicators. They may be grouped as those concerned with income, employment, education and health. There are a number of important omissions, including housing and homelessness.
8
In some NAPincl, Member States gave values for different dimensions disaggregated by gender. Figure 3 is taken from the Finnish NAPincl. In the employment field we would expect the labour market disadvantage of women to be reflected and the unemployment rate for women is indeed higher in Finland than that for men (although the long-term unemployment rate is somewhat lower) and women are disproportionately represented on employment programmes. For all other indicators, however, women were below men. In some cases, men are indeed more vulnerable. As is noted in the Joint Report, this is not typically reported in the NAPincl: “most Member States mention homelessness but few report that men comprise the majority” (Joint Report, page 75). An exception is Denmark, which notes that “80% of all homeless people are men ... fewer women fall completely through the social safety net.” (page 24). Although it should be noted that homeless women may be more vulnerable. Men may be over-represented among alcoholics and those being treated for drug abuse. A group of particular concern is that of unemployed young people who are not in education or training. As Figure 3 shows for Finland, young men may be disadvantaged more than young women. The UK NAPincl refers to “boys who have been in foster homes or in care [being] more likely to be unemployed as adults” (page 22). The Irish NAP notes that the educational levels of young women are slightly ahead of their male counterparts (page 20). Figure 4 uses the information from the Joint Report (page 198) on low education (specifically the share of the total population of 18-24 year-olds with lower secondary education or less, not attending education or training) in the year 2000. In Germany, Luxembourg, Austria and the UK the female rate is above that for men; in all other Member States the female rate is lower. In other cases, the lower score for women has to be interpreted carefully. Lower mortality and better health has to be seen in relation to what would otherwise be expected. Women may have better health than men but still be disadvantaged in terms of access to medical care. Moreover, as argued by Atkinson et al (2002), mortality as such is not an indicator of exclusion. It is not mortality as such that concerns us but differential mortality according to socio-economic or other characteristics. One country may have higher mortality than another, on account of dietary, smoking or other behavioural differences, but this does not necessarily imply a problem of social inclusion within that country. This suggests that we should look at mortality (or health) differentials by different socio-economic groups. Atkinson et al (2002, page 155) proposed as an indicator the ratio of the proportions in the bottom and top quintile groups (by equivalised income) of the population aged 15 and over who classify themselves as in a bad or very bad state of health. It would be possible to look at this indicator by gender. This would not however allow the gender dimension to be identified without further assumptions. The health differential might be the same for men and women but women might still be disadvantaged relative to men. In symbols the health status (larger numbers are bad) of men in the top class (1) and bottom class (5) is h1M = h* and h5M = h* + d5 so that subtraction gives the differential class effect (d5). Whereas for women we have w1M = w* + d1w and w5M = w* + d5 + d5w
9
where diw shows the additional disadvantage of women. Comparing the differential health status for men and women only provides a measure of the gender effect if that effect is limited to the bottom social class (quintile). In other words the identifying assumption is that upper income group women are treated like men. One purpose of considering a wider definition of social exclusion is to bring out the inter-dependence between different dimensions, including the degree of overlap. How far do women suffer multiple deprivation: i.e. are excluded on more than one dimension? Can one identify a systematic pattern of disadvantage that reveals itself across different indicators? The Danish NAP concludes that “a gender and equality perspective may lead to new knowledge about the social exclusion mechanisms” (page 24). Finally, it should be noted that the choice of indicators has in itself an impact on the picture we obtain of gender inequality. This applies both to the definition of indicators for a particular field of concern and to the choice of fields. For instance, an indicator of involuntary part-time work, where people are constrained to work part-time not full-time, should accompany the unemployment rate. Conclusion There is need for a systematic analysis of the gender dimension of social exclusion. All major indicators should be broken down by gender. The Joint Report is a step in the right direction, as the indicators given there are disaggregated in this way, but they did not cover all the indicators in Table 1. Disaggregation on its own is not however enough. The reasons for differences between men and women need to be addressed. We need to consider how far multiple deprivation reflects a general pattern of women’s disadvantage. The choice of indicators needs to be reconsidered from a gender mainstreaming perspective.
3 Gender Mainstreaming, Policy and the Role of EUROMOD What can EUROMOD contribute to a gendered analysis of the impact of policy initiatives? The starting point is that the evaluation process needs micro-based modelling. Aggregate relationships have their place, but models that treat the impact on individual households play an essential role. EUROMOD is a EU-wide model of households, constructed with the aim of evaluating changes in tax and benefit policy. It is a pilot exercise, which has served to demonstrate the feasibility of constructing a model covering 15 countries, with very different tax and social protection systems, and without a single common data source. As such, it has set itself limited objectives. It is a very useful tool but it is not allpurpose. It makes strong assumptions about behavioural response, and this limits what can be said. EUROMOD simulates a wide variety of policy instruments, including (a) income taxes, local and national, (b) social insurance contributions paid by employees, employers and the self-employed, (c) family benefits, (d) housing benefits, and (e) social assistance and other income-related benefits. There are however other taxes and benefits which are not modelled,
10
including capital and property taxes, and real estate taxes. In some cases, due to the limitations of the input data, it has not yet been possible to model pension and survivor benefits, other contributory benefits, and disability benefits. This means that there are certain classes of policy action that cannot be simulated. Further reasons why we cannot model policy initiatives are that there are attached conditions that cannot be verified in EUROMOD or that the policy is restricted to groups of the population that cannot be identified in EUROMOD. EUROMOD can be used to calculate the implications for disposable incomes. This allows direct calculation of three of the primary indicators in Table 1: (1) proportion below 60% median, (2) ratio of top quintile share to bottom quintile share, and (4) median poverty gap. It does not allow calculation of poverty persistence (indicator 3), since the model does not contain data on previous income. Given that the labour market behaviour is assumed fixed, the model cannot at present predict changes in the labour market indicators (indicators 5-7) or the proportion in education (indicator 8). One important feature of EUROMOD is that it allows different levels of analysis. The income data are assembled at the household level, but can be used to construct distributions for smaller units, such as the nuclear family, or for the individual. From a gender perspective, a distribution that distinguishes between women and men is obviously central. In order to arrive at such a distribution, assumptions have to be made about the sharing of income within the household: i.e. how far a particular source received by one individual benefits others. The mother may receive the child benefit but not be the beneficiary. EUROMOD supports a range of assumptions. It allows different intra-unit assignments of benefits and taxes. This means that we can estimate not only the proportion of women living in households whose total income puts them at risk of financial poverty but also the proportion of women whose calculated individual income puts them at risk of poverty measured in terms of individual income. EUROMOD Analysis of Gender-Oriented Policies Among the ways in which EUROMOD can be used are: • Examination of the implications of gender mainstreaming policies on employment for the indicators of social exclusion. How for example would an increase in female employment affect the levels of financial poverty? • Simulation of the differential impact by gender of policy changes. How far for example would the higher poverty rate for women than for men aged 65+ be reduced by a common (non-means-tested) minimum pension guarantee for all Member States? Among the measures that could be considered are (i) a reduction in social security contributions by employees for low paid workers (referred to in the NAPincl of Belgium, Denmark, Greece), (ii) income tax reductions for the low paid (Belgium, Denmark, Germany, Ireland, Luxembourg, Netherlands, UK), (iii) in work benefits (France, UK), and (iv) child benefits/tax credits (Greece, Ireland, Luxembourg, Portugal, Finland, Sweden, UK). • Simulation of the effect of policy changes on indicators of individual outcome. As noted earlier, it is possible to examine the impact on the distribution of individual incomes, given a set of assumptions about income sharing.
11
Conclusions The first round of NAPincl lacked a comprehensive analysis of the gender dimension of social exclusion. The economic disadvantage of women, particularly with regard to the labour market, is a theme that recurs in the NAPs, but is not systematically investigated. All major indicators should be broken down by gender, and in this respect the Joint Report is a step in the right direction. Disaggregation on its own is not however enough. We need to consider how far multiple deprivation reflects a general pattern of women’s disadvantage. The choice of indicators needs to be reconsidered from a gender mainstreaming perspective. The reasons for differences between men and women need to be addressed. EUROMOD can contribute to a gendered analysis of the impact of policy initiatives, since the starting point of the policy analysis process needs micro-based modelling. Aggregate relationships have their place, but models that treat the impact on individual households play an essential role.
12
Table A National Action Plans: Gender Dimension Belgium Population 10.2 million Low wages, part-time work, unequal share of career breaks, precarious work: all these characteristics of female employment are identified as elements to be combated. A list of measure taken by the Minister for equality is given in “le plan stratégique pour l’égalité des chances”. Gender is specifically noted with regard to the risks of lone parents not in receipt of alimony, and the under-representation of women on training programmes. When asked about going to the movies, attending a cultural meeting, going into a bar, a restaurant, or dancing, 7% of men tell that they never had such an activity during previous year compared with 10% for the whole population (the percentage for women is not given but is evidently higher) (p.23). The actions/policies under Objective 1.1 are: • Employment policy in its different axes has to be implement in order to promote equality between men and women (p.24) • Maribel reduction of social contributions in the social economy mainly benefit to women (64.6%) (p.26) • Reform of the income tax system: a tax credit is implemented to help low wage earners, which could increase female participation in the labour market (p.26) • Premium to lone parents: paid to lone parents long term unemployed taking a job (mostly women) (p.26) • “Plan avantage à l’embauche”, plan +1, +2, +3: reduction of employers’ social contribution for women re-entering the labour market: (p.26) • Child Care facilities: essential condition for women’s labour market participation: o Tax deduction o Increasing in the availability (p.28) • To create a policy of equal opportunities for men and women (p.29): Objective 2. • Pensions: to promote equality: the minimum age to get a pension will be progressively increased to 65 for women (instead of 60) • Alimony: the government will take measures in order to facilitate the recovery of alimony • Health: women in poor families are less likely to benefit from cancer screening; policies will be undertaken in order to facilitate access. Indicators None of the indicators is presented by gender.
13
Denmark Population 5.3 million The NAP notes that “Danish social policy supports a very high activity rate for men and women alike ... the majority of Danish pre-school children are looked after in public daycare facilities”(page 8). Parental leave to take care of young children has been available since 1994; 93% of the parents on parental leave are women (page 18). The report differential risk of marginalisation for men and women: “to some extent, women and men are excluded from the labour market for different reasons, and to some extent they also react differently to the situation ... 80% of all homeless people are men ... fewer women fall completely through the social safety net ... However there are signs of a shift in the described gender-related patterns among socially marginalized people. Among the 18-21 year olds, the distribution by sex is more even.” (page 24). A new Equal Opportunities Act was adopted in 2000, establishing statutory gender mainstreaming. It requires all public authorities to incorporate equality in their planning and administration (page 41). Indicators Some indicators are broken down by gender. Germany Population 82 million. The NAPincl gives especial weight to reducing unemployment and notes that the number in employment, particularly women, has grown since 1998 (page 6). The NAP refers to a number of measures as being of particular benefit to women: legislation to help public sector employees who work part-time for family reasons, increased pension entitlement for those who have worked part time, measures to help the victims of domestic violence. When discussing the position of young people with non-German mother tongue, “particular attention is focused on the difference between the sexes as regards the opportunities open to them” (page 40).
Greece Population 10.5 million The NAP in its account of the “transition to a new model” (page 4) notes that there has been an increase in the last few years in the number of women participating in the labour market and higher education. Three-quarters of net inflows into employment in 1990-1999 were women (page 25). It recognises however that women “face notable problems of equal access to the labour market” (page 10). At younger ages, this is due to “problems related to the reconciliation of work and family life”, but older women “face two problems: an increasing number will be living alone, perhaps due to a divorce ... and access to the labour
14
market is problematic due of their skill facing obsolescence [and other] obstacles and inequities” (page 10). They note that public policy may be counter-productive: “early retirement incentives, by urging women to early retirement with few years of contributions, often press them to depend on other family members. In the event of family break-up, this may lead to dramatic social problems” (page 10). The NAP has a chart (page 13) summarising the main actions, including those to help women reconcile work and the family. The measures listed include day nurseries, children’s summer camps and home assistance. The NAP sets out (page 25) proposals to mainstream the gender approach, including use of ESF funding, training of employment counsellors, education and training programmes for women on new technologies. Under Objective 1.1, the government are introducing an employee contribution subsidy for workers earning the minimum wage; and reduction of employer contributions of 2% for low paid workers. These measures “are expected to especially benefit women” (page 25). Indicators Poverty figures are given for men and women. The Welfare Map is to give figures broken down by gender.
Spain Population 39 million Women are presented as a risky group concerning employment, training, and income (p.7, 8). The negative impact of gender on different precarious situations has to be taken into account (p.9). Disadvantaged women include those suffering domestic violence, the unemployed, and those without diploma. Actions/policies Objective 1.1 General: taking gender into account in employment policies • To promote new work arrangements for women with dependants (time flexibility, home working) (p.11) • To develop neighbourhood services (social economy) to facilitate female labour force participation (p.11) Objective 1.2 • Pensions: take measures to avoid discrimination against women in rural areas • Training/education: to promote the participation of women without qualification (ALBA) Objective 3. • Plan to combat domestic violence against women (p.27) • Measure to help handicapped women (p.29)
15
•
A whole paragraph is devoted to the improvement of the situation of disadvantaged women presenting different measures and plans (p.30) o to promote social and professional inclusion, o to help women in rural areas o to help prostitutes o to insure access to social protection for lone women o to combat domestic violence against women: by education, women’s refuges, meeting points.
Indicators The Spanish NAP gives basic data for each objective. Footnote p.40: all indicators should be disaggregated by sex
France Population 59 million Women have higher unemployment rates. Higher risk of poverty for low educated young women Actions/policies Objective 1.1 ARAF: help for women re entering the labour market Objective 2. family solidarity: the place of father in child raising, the right of child to be educated by both parents (fatherhood) Objective 3. TRACE: this programme will be more targeted on low educated women Indicators Special attention will be paid to the presentation of indicators by sex (p.46) Proposed as indicators: Percentage of people who are unable to work due to care (children or other dependants) Child care: Number of places in the public sector School attendance rate of children less than 3 years olds.
Ireland Population 3.7 million The National Anti-Poverty Strategy was established in 1997, and was under review at time of NAPincl. Among the new themes in the review is women’s poverty (pages 3 and 10).
16
The National Development Plan (NDP) 2000-2006 has a gender equality perspective (page 4) and the issue is one of the responsibilities of the NDP Equal Opportunities and Social Inclusion Co-ordinating Committee (page 11). There is a Gender Equality Unit (page 18), which provides training on gender mainstreaming to government officials. Initiatives in the area of childcare are “critical to supporting enhanced participation of females in the labour market” (page 7). The NAPincl discusses the employment situation and the progress made towards achieving the Lisbon employment targets. The NAPincl does not refer to new policy but describes recent policy initiatives under the NAPS and the national agreement, the Programme for Prosperity and Fairness (PPF), and under the NDP. The Equality for Women Measure of the NDP seeks to develop initiatives in training for women, encouraging entrepreneurship and career development, gender proofing of personnel practices, supporting women in decision-making and family friendly employment (page 30). There are programmes to address domestic violence (Annex I). There is an Advisory Committee on the Role of Women in Agriculture (Annex L). Indicators There is a commitment to develop a system for the collection of disaggregated data for women. Italy Population 58 million The traditional Italian welfare model is a male breadwinner one, women bear the responsibility of child raising, dependant care, and domestic activities. This role of women is a potential source of risk for them. Women’s labour force participation is needed for social inclusion. Horizontal and vertical segregation, wage inequalities, the inequality in the sharing of domestic work, and the lack of childcare facilities are all criticised. There is an over representation of women among the disabled. There is concern about female immigrants who are victims of human trafficking. “Piano nazionale occupazione” contains different measures to combat discrimination
Actions/policies Objective 1.1 • the new body for employment will have to promote female activity • fiscal deduction of child care costs • premium for lone parents • marginalisation of immigrated persons: high risk of sexual traffic for women • Valorisation of family responsibility: maternity allowance, parental leave
17
Objective 2. • family solidarity: creation of different services and help to promote conciliation Indicators Different data concerning poverty, education, literacy, Internet and PC access are given by gender.
Luxembourg Population 0.4 million There is need to extend the child care facilities. Women willing to re-enter the labour market after child raising encounter difficulties due to the competition of foreign workers and the lack of child care facilities Among the policies for fighting social exclusion is that for the promotion of women (p.29) o Promotion of equality in all measures taken o Priority o Conciliation o Sharing of pension rights between men and women o Valorisation of child raising periods for the calculation of pensions Actions/policies Equality as a permanent concern Objective 1. • Raising of women/men employment in the sectors where they are under represented • Child care facilities to help the conciliation of work and family life Indicators Data are presented by gender
Netherlands Population 15.7 million The NAPincl notes thre sharp increase in the labour market participation rate of women, from 39% in 1990 to 52% in 2000 (jobs of 12 hours per week or more), although the rate is still 25 percentage points below that for men. The government aims to raise the activity rate to 65% by 2010. Women aged over 65 form the largest group with long-term dependence on minimum income. The government has taken measures to combat domestic violence.
18
The first Emancipation Monitor (Emancipatiemonitor 2000) has focused on employment, care and income, on education, on power and decision-making, and on violence against women. Indicators Some indicators are broken down by gender.
Austria Population 8.1 million The NAPincl sets out under Objective 1.1 its aim to promote education for women: “the gender-specific segmentation of the labour market and the less favourable job opportunities for women in general call for specific measures to promote the education and training of women and girls” (page 6). The Action Plan 2003 gives prominence to women in technology. The NAP refers to the better reconciliation of family and working life. The childcare allowance introduced in 2002 “should help remove the existing employment barriers for parents ... It is expected that the 30% rate of re-entry following parental leave will rise considerably” (page 9). It is also hoped that “the new arrangements should contribute to reducing the earnings gap between men and women. The new rules should give men an incentive to spend more time on childcare ... If childcare is shared more evenly between parents, the career opportunities of mothers are also improved” (Page 19). The government seeks to encourage family-friendly employment (page 10). The European Social Fund is being used to support equal opportunities policies (page 12). The Labour Market Service’s objectives of February 2001 provide that 50% of the budget should be devoted to women under the heading of active employment market policy (page 13). The government plans to enable women who have no independent pension entitlement to acquire their own pension, recognising part of the childcare period as a pension contributory period (page 20). The government has taken measures to address the non-payment of maintenance by absent fathers. In the event of default, advances are paid by the state. The 1997 Violence Protection Act created the legal prerequisites for action against domestic violence.
Portugal Population 10.0 million
19
The NAP notes that even if women are a minority amongst homeless, drug consumers, etc…. they are particularly vulnerable if they belong to these groups (p.6). The NAP stresses the importance of equal access to family life for men. Actions/policies Objective1.1 • •
• • •
To promote equality between men and women in access to employmentby strengthening the rights to men to be considered as working fathers (p.13) To include in the social organisation and in enterprise culture the idea that conciliation between work and family is a right and a duty for all workers (men and women) and a responsibility for the enterprises To develop child care facilities and to promote training and social inclusion for all family members To combat wage inequalities, to prevent child labour and to promote female participation To inform managers about equality concerns
Aims: • increase training participation ensuring equal participation of both sexes • raise to 26000 the number of people in apprenticeship encouraging the participation of the under represented sex Objective 2 • To take into account the gender dimension in all measures proposed including in access to new information technologies (equal participation of men and women) (p.29) • To facilitate female integration in the labour market and men’s participation in family life • developing care networks, doubling the number of child care facilities (crèches 2000) Objective 3. •
to develop networks in order to help women victims of violence
Good practices • Crèches 2000 • Strengthening men’s rights as working fathers to favour reconciliation between work and family. Indicators None of the indicators is presented by gender.
Finland Population 5.1 million
20
The NAPincl makes relatively little explicit reference to gender issues. It notes that one of the benefits of a largely individual-based system of social protection is “the enhancement of gender equality” (page 21). Indicators Indicators for health, unemployment, employment promotion, unemployed schoolleavers, suicides and alcohol and drug-related hospital treatment are broken down by gender.
Sweden Population 8.8 million The Swedish NAPincl opens with the statement that “one of the main tasks of employment policy is to avoid a gender-segregated labour market and contribute in other ways to greater equality between men and women” (page 1). Changes to the parental insurance system mean that the proportion of fathers taking parental leave increased from 7% in 1990 to over 12% in 2000. The NAP identified women as particularly affected by health problems, noting that the divorce rate following a period of illness is higher for women than for men (page 6).
UK Population 59 million The analysis of poverty is seen in life-cycle and inter-generational terms: “breaking the cycle of deprivation and ending the inter-generational transmission of disadvantage from parent to child must lie at the heart of any effective strategy for tackling poverty and social exclusion” (page 15). Much of the emphasis is on helping children, but it is recognised that “there is a significant gender dimension to poverty – women constitute a higher proportion of adults in poverty and are more likely to be persistently poor. Women are particularly likely to have low incomes at key stages of their life cycle e.g. lone mothers and single women pensioners” (page 15). The NAPincl reports exclusively on existing policies, particularly those concerned with active labour market policies and tax and benefit reform.. Employment is seen as the key, and this has a clear gender dimension, as in the National Childcare Strategy: “development of effective child care strategies will have a crucial impact on women’s lives by increasing their opportunities to enter the labour market and build up lifelong incomes” (page 9). Among the other policies where a gender dimension is identified is the introduction of a National Minimum Wage, benefiting some 1.3 million workers from April 1999; “in narrowing the gender pay gap the minimum wage has had the greatest effect on women’s pay since the Equal Pay Act 1970” (page 8). The pension reforms “are of particular benefit to women [supporting] those who have suffered in the past from broken work records ... the stakeholder pension will be of particular benefit to women as it will provide a flexible, secure
21
and value for money second pension option ... those who take career breaks will be able to stop and start contributions without penalty” (page 21). There is extensive discussion of teenage pregnancy, where the UK has one of the highest rates in the EU. An example is given of an ESF-supported women’s ICT training centre in East London. Indicators Some indicators are broken down by gender.
22
TABLE 1 Primary Indicators Agreed by European Union December 2001 1. Percentage of individuals living in households with low incomes (below 60% of the national median equivalised income); 2. Persistent financial poverty; 3. Depth of financial poverty; 4. Ratio of income of top 20% to that of bottom 20%; 5. Coefficient of variation of regional employment rates; 6. Long-term unemployment rate; 7. Percentage of people living in jobless households; 8. Early school leavers not in further education/training; 9. Life expectancy at birth: 10. Self perceived health status by income level.
23
References to NAPincl Government of Belgium, 2001, Plan d’Action National Inclusion Sociale, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html Available in French and Dutch. Government of Denmark, 2001, Denmark’s National Action Plan to Combat Poverty and Social Exclusion (NAPincl) 2001/2003, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Danish and English. Government of Federal Republic of Germany, 2001, National Action Plan to Combat Poverty and Social Exclusion (NAPincl) 2001-2003, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in German and English. Government of Finland, 2001, National Action Plan Against Poverty and Social Exclusion, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Finnish and English. Government of France, 2001, Plan national d’action français contre la pauvreté et l’exclusion sociale, EU website europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in French and published by La documentation Française, Paris, 2001, ISBN 2-11004906-5. Government of Greece, 2001, National Action Plan for Social Inclusion, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Greek and English. Government of Ireland, 2001, National Action Plan Against Poverty and Social Exclusion (NAPincl) 2001-2003, EU website europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in English. Government of Italy, 2001, Piano nazionale per l’inclusione 2001, EU website europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Part I available in Italian. Government of the Netherlands, 2001, Dutch National Action Plan Against Poverty and Social Exclusion, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Dutch and English. Government of Portugal, 2001, Plan National d’Action pour l’Inclusion, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html.
24
Available in Portuguese and French. Government of Republic of Austria, 2001, National Action Plan to combat povewrty and social exclusion, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in German and English. Government of Spain, 2001, Plan national d’Action du Royaume d’Espagne pour l’Inclusion Sociale juin 2001-juin 2003, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Spanish and French. Government of Sweden, 2001, Sweden’s Action Plan Against Poverty and Social Exclusion, EU website europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in Swedish and English. Government of the United Kingdom, 2001, United Kingdom National Action Plan on Social Inclusion 2001-2003, EU website, europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in English. Grande-Duché de Luxembourg, 2001, Plan National d’Action Pour L’Inclusion Sociale, EU website europa.eu.int/comm./employment social/news/2001/jun/napsincl2001 en.html. Available in French.
25
References Atkinson, T, Cantillon, B, Marlier, E and Nolan, B, 2002, Social Indicators: The EU and Social Inclusion, Oxford University Press, Oxford. European Anti-Poverty Network (EAPN), 2001, EAPN Synthesis Report on the 2001-2003 National Action Plans on Social Inclusion, EAPN, Brussels. European Commission, 1998, A Guide to Gender Impact Assessment: Equality Between Women and Men, Office for Official Publications of the European Communities, Luxembourg. European Commission, 2000, “Structural Indicators”, Communication from the Commission, COM (2000) 594 final. European Commission, 2001, “Draft Joint Report on Social Inclusion”, Communication from the Commission, COM (2001) 565 final. European Commission, 2002, The Lisbon Strategy – Making Change Happen, Communication to the Spring European Council in Barcelona, Brussels, COM(2002) 14 final. European Commission, 2002a, Presidency Conclusions Barcelona European Council, Brussels. Eurostat, 1999, European Community Household Panel (ECHP): Selected Indicators from the 1995 wave, Office for Official Publications of the European Communities, Luxembourg. Marlier, E, and Ponthieux, S, 2000, “Low-wage employees in EU countries”, Statistics in Focus, Theme 3, 11/2000. Mejer, L and Linden, G, 2000, “Persistent income poverty and social exclusion in the European Union”, Statistics in Focus Theme 3 13/2000. Mejer, L and Siermann, C, 2000, “Income poverty in the European Union: Children, gender and poverty gaps”, Statistics in Focus, Theme 3 12/2000. Rees, T, 1995, “Follow-up of the White Paper on Growth, Competitiveness and Employment, Equality into Education and Training Policies”, European Commission, Directorate-General V, Employment Task Force V/3, V/5538/95-EN. Ruxton, S and Bennett, F, 2002, Developing a coherent approach to Child Poverty and Social Exclusion across Europe, Eurnoet, Brussels. Social Protection Committee, 2001, Indicators Sub-Group: Report from the Chairman to the SPC, Brussels.
26
Figure 1 Relative Incidence of Low Monthly Wages men
2.5
women
Relative incidence
2
1.5
1
0.5
EU 13
K U
P
A
L N
L
I
L IR
F
E
EL
D
D
K
B
0
Figure 2 Relative EU Poverty Rates All 1.7
65+
1.6
Persistent
Women relative to men
1.5 1.4 1.3 1.2 1.1 1 0.9 B
DK
D
EL
E
F
IRL
I
L
27
NL
A
P
UK
EU13
K
I
L
28
L
EU
K
15
U
S
FI N
P
A
N
L
IR
F
E
EL
D
D
B
U
U
ne m
ne m
pl oy
pl oy
ot
ot
Tr ea
n
n
tm en tf
or
ol
ug s
co h
dr
al
-2 4
-1 9
io n
t
id es
20
15
m ot
Su ic
io
io
or
ca t
ca t
en tf
ed u
at m
in
ed u
oy m en
t
y
y
oy m en
pr o
em pl
en t
un
oy m
in
Tr e
ed /n
ed /n
Em pl
LT
m pl
ilit
th
ta lit
m ob
he al
M or
ite d
U ne
Li m
Po or
%
Figure 3 Finland NAPincl: Women as % Men
180
160
140
120
100
80
60
40
20
0
Figure 4 Female/Male Ratio of Low Educated Young People
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
